All Topics
All Topics
Technology
Technology
Design
Design
Programming
Programming
Science
Science
News
News
Gaming
Gaming
Entertainment
Entertainment
Business
Business
Finance
Finance
Sports
Sports
Health
Health
Food
Food
Travel
Travel
Art
Art
Music
Music
Books
Books
Education
Education
Politics
Politics
Personal
Personal
No algorithm. No AI slop. No ads. Just RSS. Pro-human. Indie writers. Real journalism. Open web. Chronological. Hand toasted.

Research Study: AI Coding Assistants' Tool Recommendations Analysis

By

tin7in

3mo ago· 2 min readenInsight

Summary

A research study analyzing AI coding assistants' tool recommendations by testing Claude Code on real repositories 2,430 times. The study examines what development tools Claude Code, Codex, and Cursor actually recommend when given open-ended questions without tool names in prompts. Key findings show Claude Code prefers custom/DIY solutions over pre-built tools, with an 85.3% extraction rate across 3 models, 4 project types, and 20 tool categories. The research includes a head-to-head benchmark comparison of OpenAI Codex against Claude Code with 1,470 responses across 12 categories.

Key quotes

· 4 pulled
We pointed Claude Code at real repos 2,430 times and watched what it chose. No tool names in any prompt. Open-ended questions only.
The big finding: Claude Code builds, not buys. Custom/DIY
3 models · 4 project types · 20 tool categories · 85.3% extraction rate
We benchmarked OpenAI Codex head-to-head against Claude Code — 1,470 responses across 12 new categories.
Snippet from the RSS feed
What dev tools do Claude Code, Codex, and Cursor actually recommend? Systematic research across real repos and model generations.

You might also wanna read